Date: Sat, 14 Mar 2015 10:51:39 -0700 From: Tim Kientzle <tim@kientzle.com> To: Brett Wynkoop <freebsd-arm@wynn.com> Cc: freebsd-arm <freebsd-arm@freebsd.org> Subject: Re: BeagleBone slow inbound net I/O Message-ID: <44FB8364-1310-4FBF-BA79-1B8837DE1C79@kientzle.com> In-Reply-To: <20150314031542.439cdee3@ivory.wynn.com> References: <20150311165115.32327c5a@ivory.wynn.com> <89CEBFCA-6B94-4F48-8DFD-790E4667632D@kientzle.com> <20150314031542.439cdee3@ivory.wynn.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> On Mar 14, 2015, at 12:15 AM, Brett Wynkoop <freebsd-arm@wynn.com> = wrote: >=20 > [wynkoop@beaglebone ~]$ sysctl dev.cpsw | grep -i error > dev.cpsw.0.stats.RxCrcErrors: 262 > dev.cpsw.0.stats.RxAlignErrors: 231 > dev.cpsw.0.stats.CarrierSenseErrors: 0 > [wynkoop@beaglebone ~]$=20 >=20 > So we can see climbing errors. I am not sure how this compares to the > results of others. The above was during the first few minutes of a > buildworld from an nfs share. Your Ethernet hardware is detecting a lot of low-level network errors, which generally indicates hardware problems in your network: a bad cable maybe, or a dying switch or bad power supply. In any event, NFS throughput degrades rapidly on a poor network, so this probably explains your NFS slowness. There are a lot of error-related counters that aren=E2=80=99t called = Errors, by the way. In particular, the various collision and overrun counters don=E2=80=99t actually have =E2=80=9CError=E2=80=9D in the = name. On my BeagleBones, the only non-zero error counter I generally see is the RxStartOfFrameOverruns. Tim
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44FB8364-1310-4FBF-BA79-1B8837DE1C79>